Terahertz imaging of subjects with concealed weapons

7/29/2019 Terahertz imaging of subjects with concealed weapons

1/13

Terahertz imaging of subjects with concealed weapons

Jason C. Dickinson*a, Thomas M. Goyettea, Andrew J. Gatesmana, Cecil S. Josepha,Zachary G. Roota, Robert H. Gilesa, Jerry Waldmana, and William E. Nixonb

aSubmillimeter-Wave Technology Laboratory, University of Massachusetts Lowell, 175 CabotStreet, Lowell, MA 01854

bU.S. Army National Ground Intelligence Center, 2055 Boulders Road, Charlottesville, VA 22911

ABSTRACT

In response to the growing interest in developing terahertz imaging systems for concealed weapons detection, theSubmillimeter-Wave Technology Laboratory (STL) at the University of Massachusetts Lowell has produced full-bodyterahertz imagery using coherent active radar measurement techniques. The proof-of-principle results were readilyobtained utilizing the compact radar range resources at STL. Two contrasting techniques were used to collect theimagery. Both methods made use of in-house transceivers, consisting of two ultra-stable far-infrared lasers, terahertz

heterodyne detection systems, and terahertz anechoic chambers. The first technique involved full beam subjectillumination with precision azimuth and elevation control to produce high resolution images via two axis Fouriertransforms. Imagery collected in this manner is presented at 1.56THz and 350GHz. The second method utilized afocused spot, moved across the target subject in a high speed two dimensional raster pattern created by a large two-axispositioning mirror. The existing 1.56THz compact radar range was modified to project a focused illumination spot onthe target subject several meters away, and receive the back-reflected intensity. The process was repeated across twodimensions, and the resultant image was assembled and displayed utilizing minimal on-the-fly processing. Imagery at1.56THz of human subjects with concealed weapons are presented and discussed for this scan type.

Keywords: Terahertz, imaging, concealed, scanner, transceiver

1. INTRODUCTIONThe terahertz region of the spectrum is generally considered to be the frequency range between 0.3 THz and 3 THz. Thewavelengths in this region therefore vary from around 1 millimeter to 0.1 millimeter. These long wavelengths areknown to penetrate common clothing materials, and are short enough to produce substantial detail of items concealedbeneath clothing. This study is intended to provide a demonstration of the phenomenology of terahertz imaging and adiscussion of the difficulties encountered. Since the Expert Radar Signature Solutions (ERADS) program currentlyoperates indoor compact radar ranges in this frequency region, standard active radar techniques are applied to theproblem of concealed weapons detection.

Since 1981 the Submillimeter-Wave Technology Laboratory (STL) at the University of Massachusetts Lowell hasperformed radar signature measurements of physical scale models of tactical vehicles using submillimeter-wavetransceivers. The development and application of these systems has been funded by the U.S. Army National GroundIntelligence Center (NGIC) under the ERADS program. Currently STL and NGIC have six compact radar ranges

operating at frequencies up to 1.6 terahertz. The terahertz sources and receivers, anechoic chambers, and computercontrolled positioning apparatus available at STL made the data collection for concealed weapons straightforward,requiring only minor changes to the existing 1.56THz and 350GHz systems1. Previous work described high resolution,active transmission, terahertz imagery of a clothed mannequin with a concealed weapon, measured in the ERADS/STL1.56THz and 350GHz compact ranges using azimuth/elevation (Az/El) collection techniques. A brief review andadditional results will be presented.____________________________________

* Correspondence: Email: [email protected]; Telephone: (978) 934-1381; Fax: (978) 452-3333


2/13

Report Documentation PageForm Approved

OMB No. 0704-0188

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and

maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,

including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington

VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing t o comply with a collection of information if it

does not display a currently valid OMB control number.

1. REPORT DATE

MAY 20062. REPORT TYPE

3. DATES COVERED

00-00-2006 to 00-00-2006

4. TITLE AND SUBTITLE

Terahertz imaging of subjects with concealed weapons

5a. CONTRACT NUMBER

5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

University of Massachusetts Lowell,Submillimeter-Wave Technology

Laboratory,175 Cabot Street,Lowell,MA,01854

8. PERFORMING ORGANIZATION

REPORT NUMBER

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITORS ACRONYM(S)

11. SPONSOR/MONITORS REPORT

NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT

Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES

The original document contains color images.

14. ABSTRACT

15. SUBJECT TERMS16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF

ABSTRACT18. NUMBER

OF PAGES

12

19a. NAME OF

RESPONSIBLE PERSONa. REPORT

unclassified

b. ABSTRACT

unclassified

c. THIS PAGE

unclassified

Standard Form 298 (Rev. 8-98)Prescribed by ANSI Std Z39-18


3/13

Due to the stringent positioning requirements of Az/El data collection techniques, a system which was better suited todemonstrate the measurement of human subjects was considered, and a prototype system was investigated, utilizing theavailable 1.56THz transceiver system. The prototype system employed a raster style spot scan implementation, where afocused spot is rastered across the target subject at high speed. A description of this system and imagery data will bepresented.

In separate terahertz materials scaling projects, an extensive library of materials transmission spectra were collected fora broad range of terahertz frequencies and compiled. A sampling of the data is available online at the STL web sitehttp://stl.uml.edu and in reference (2) of this conference.

2. THE 1.56THZ TRANSCEIVER SYSTEM

The 1.56THz transceiver system at STL uses two carbon dioxide lasers paired individually with two far-infrared lasers.All four units are designed for extremely long term power stability over periods of several days. The transceiver provedto be an ideal test-bed for collecting measurements of the clothed mannequin, in order to gauge the overall feasibility ofusing terahertz to penetrate clothing and perhaps reveal concealed metallic weapons. The system has excellentsensitivity, which is obtained through a combination of fairly high transmitter beam power (100+milliwatts) and

sensitive heterodyne receivers.

A sensitivity of approximately 10-19 W/Hz is typical for the 1.56THz systems receiver. The receiver plays a crucial rolein measuring the low-level signal back-reflected from the subject at useful standoff distances. Corner-cube mountedSchottky diode mixers provide sensitivity for heterodyne detection in the 1-2 THz range. The heterodyne nature of thereceivers allows extraction of the amplitude and phase of the backscattered signal, which is essential for the complexFourier post-processing of the Az/El measurements.

Transmitting beams over moderate distances of 10-30 meters at frequencies in the terahertz region can be problematicas this frequency range is highly susceptible to water vapor absorption. At 1.56, THz there is a narrow window of lowwater vapor absorption, where the effect is limited to approximately 0.1-0.2 dB of loss per foot for compact rangehumidity conditions near 20% RH (relative humidity).The output power, over 100 milliwatts, in the transmitter laserbeam allows sufficient power to easily transmit through the nearly 80 foot round trip distance to and from the subject.

3. THE AZ/EL MEASUREMENT TECHNIQUE

The Az/El technique utilizes full-beam illumination of the target and achieves high resolution images by viewing thetarget through a 2D angular aperture similar to synthetic aperture radar1. Using data collection techniques commonlyemployed in all ERADS compact radar ranges to generate radar backscatter imagery, a clothed mannequin wasrotationally incremented, following a precise trajectory along azimuth and elevation axes. Complex Fourier transformpost-processing along the orthogonal azimuth and elevation axes orient pixel values in vertical and horizontal cross-range. The post-processed images created have a resolution of around 1.1 mm per pixel. Figures 3.1 through 3.3 showAz/El images collected in the 1.56THz and 350GHz systems. Data collection time for a full 360 degrees of imagery wastypically 3-5 hours, with individual image frames (as shown) corresponding to 1-2 minutes per frame. A completedescription of this has been presented in reference (1).


4/13

Figure 3.1. Azimuth/elevation collection of a clothed mannequin with a concealed, metal tie-wrap gun(inset). Left image taken at

1.56THz. Center image taken at 350GHz (dBV2 color scale). Photograph of subject as measured on right. Note thepenetration through the black plastic wrapping of the trigger handle at 350GHz. The black plastic wrap is heavilyattenuating at 1.56THz and minimally attenuating at 350GHz, allowing the handle metal to reflect brightly.

Figure 3.2. A 350GHz Az/El image (left) of a .45 Caliber handgun and a photograph of the measured gun (right) (dBV2 color scale).

Figure 3.3. A 350GHz Az/El image (left) of a roaster turkey and a photograph of the turkey as measured (right). The raw roastingturkey was measured to approximate the reflection characteristics of human skin (dBV2 color scale).


5/13

The azimuth/elevation collection technique relies heavily on the slow and controlled phase change of the complex targetas a function of the angular position change. In order to satisfy the Nyquist limit, the phase change occurring at thetarget extremities can not change more than 90 degrees (/2) of phase per data collection angular position increment orphase wrap (aliasing) will occur. This maximum allowable phase change is examined in equation (1), where theunambiguous cross-range (image width) is a function of the data collection angular increment3. For the post-processed

image resolution, equation 2 is used. Effectively the larger the angular integration, the finer the resolution of the image,and the finer the data collection, the wider a target can be. The ultimate image resolution is tied into the processedangular window width for the horizontal resolution, and the processed angular window height for the vertical resolution(equation 2).

Runamb = / 2increment (1)

Rc = / 2integrated angle (2)

The Az/El data collection capability acts as an excellent study in terahertz phenomenology of a complex target shapecomposed of multiple dielectric materials. A difficulty of the Az/El data collection method lies in the intolerance tounexpected changes in received phase. Motion outside of the predicted collection path causes phase corruption,

resulting in severe image distortion. The required precision of the data collection angles limits the Az/El measurementtype for measuring a live subject, possibly even if an array of receivers were employed. Measurements require the targetsubject to remain motionless, except for a controlled series of angular changes in its position. For a system operating at1.56THz, unpredicted movements of more than a small fraction of the 191 micron wavelength over the course of severalhours would cause image distortion. Therefore, a different method of data collection was devised for practical THzmeasurements of freestanding humans. Section 5 describes a spot scanning system that is more practical for thisapplication.

4. MATERIAL CONSIDERATIONS FOR DESIGN OF A TERAHERTZ SCANNER

Figure 4.1 shows a transmission plot of a small sampling of measurements taken at STL of clothing materials as afunction of terahertz frequencies from 300GHz to 1.5THz. Comparisons of numerous Az/El images with terahertz

materials spectra show good agreement between the Az/El clothed mannequin measurements and the materialstransmission data2. While reflectivity of metals across the terahertz region is near unity4, most clothing materials aretransmissive enough to pass some terahertz energy, whichcould make a properly designed system useful for homelanddefense applications.

Materials properties of various common clothing, building materials, and plastics have been collected over a wide rangeof terahertz frequencies. It is difficult however to perform the same type of terahertz material characterization on livetissue. In order to approximate the reflective properties of skin and tissue, a supermarket roasting turkey was measuredat 350GHz in an Az/El scan configuration. A sample image is shown in Figure 3.3. The reflection of the turkey skinappears to be similar to the reflective properties of the fiberglass mannequin. It should be noted however that theabsorption properties between fiberglass and human skin are expected to be different.

The terahertz spot scanner, described in detail in the next section, may prove to be a useful tool in performing novel

measurements of living tissue. If an optical system were redesigned to manipulate the terahertz beam across small scanareas with a small beam focus, living tissue could be examined at high speed and at close range. Pulsed terahertzradiation has been used in-vivo to show contrast between healthy skin and a basil cell carcinoma5. Use of a CW basedterahertz scanner operating at some optimal frequency may offer high speed, non-contact, high resolution imagery todetect numerous skin conditions, from cancerous tissue to burn depth.


6/13

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0

Transmittance

1.501.401.301.201.101.000.900.800.700.600.500.400.30

Frequency (THz)

LLBean warm up jacketThick wool coat100% cotton orange shirt100% cotton jacketThick sweatshirtDenim jeans100% cotton tee shirt

Figure 4.1. Plot of transmission through common clothing materials over terahertz frequencies from 0.3 THz to 1.5 THz. Apparent

feature at 800 GHz for all samples is a spectrometer artifact.

5. THE PROTOTYPE SPOT SCANNING SYSTEM AT 1.56THZ

The transition from the compact radar range setup, with full-beam illumination, to a high speed spot scan system wouldenable characterization of complex target subjects without the stringent positioning requirements. In order to study theterahertz characteristics and phenomenology of clothing materials on humans, the spot scan method was devised, and aprototype system was assembled. Figure 5.1 is a simple illustration of the difference in spot size from the full-beamillumination and the prototype spot scan beam size. The following sections are arranged to describe in general the issuesassociated with delivering a focused spot at the target subject located several meters away.

Figure 5.1. Relative sizes of the illumination beam for the full-beam system (left) and the spot scan systems (right).


7/13

5.1 Terahertz beam considerations

Optical systems utilizing terahertz beams are quasi-optical, that is, the wavelength is nearly long enough be behave asan RF signal, yet short enough to be manipulated by optical components, but the size of the optics must take diffractive

effects into account.

The distance between the final optic in the illumination beam path and the target subject is referred to as the standoffdistance. The beam emerging from the last transmit optic, usually a focusing mirror, propagates freely through the airuntil it reaches the subject. It is usually desirable to maximize this distance, in order to remotely operate a scannersystem as far away as possible from the subject that may have a concealed weapon. The common misconception for aterahertz system is that the beam can propagate across great distances as a narrow, focused beam. All types of lightexperience some form of diffraction effect as they propagate, and terahertz Gaussian beams tend to experience fairlyrapid expansion as they propagate. Figure 5.2 shows the diffraction of a 1cm terahertz beam at 300 GHz, 1.5 THz, and 3THz over a distance of 2 meters using standard diffraction calculations.

Figure 5.2. Initial Gaussian beam waist size of 10mm, propagated over 2 meters, for three frequencies, 0.3THz (red, upper),1.5THz (blue, middle), and 3THz (black, lower).

5.2Terahertz optic sizingBecause the intensity of a Gaussian beam drops off radially from the center, it does not have a crisp edge where beamenergy stops. A terahertz beam spot is subject to numerous interpretations. Figure 5.3 shows the values for severalcommon terms used to describe the size of a beam. Full width at half maximum (FWHM) is a common reference, asare beam waist () and pi omega (). In order to strike a balance between practical optical component size and the

utilization of the majority of beam power, the size of is commonly used for the minimum diameter of an optical

component in a Gaussian optical system. The diameter of contains 99% of the beam intensity, with only 17%

ripple6

. For example: A Gaussian beam with a 25mm beam waist would have a FWHM size of nearly 30mm, and inorder to satisfy would require a minimum optic diameter of 78.5 mm.


8/13

Figure 5.3. A Gaussian beam intensity profile, showing beam width definitions FWHM (black) , 1/e2 (dotted), and (blue).

5.3Terahertz Standoff issuesIn order to calculate the minimum size of the last optic in the terahertz transmission path, Gaussian beam propagationcalculations were used. Gaussian beams are subject to time-reversal, so a simple approach to calculate parameters of a

focused spot several meters away is to mentally reverse the propagation direction. If a 12mm spot FWHM (10mmwaist) at 1.5THz is desired 10 meters away, the approach would start with a 10mm beam waist and propagate it 10meters to calculate the required optic diameter. Figure 5.3 shows the 10mm beam diffraction over 20 meters. The beamwaist at 10 meters is approximately 62mm. Using the final illumination beam optic size would be ( * 63mm =)198mm diameter, approximately 7.75 inches. If a 20 meter standoff was desired, Figure 5.4 shows the beam waist at 20meters to be 125mm, approximately double the beam waist size at 10 meters. For 20 meters of standoff the optic

required would be 393 millimeters, or 15.5 inches in diameter. Table 5.1 shows the required optic diameter for severalpermutations of beam frequency, beam size, and standoff distance.

Table 5.1. Focusing optic diameter ( inmm) as a function of final beam waist size (mm) and standoff distance (meters) at 1.5THz.

5 meter 10 meter 20 meter 30 meter

5mm 198 393 801 12097.5mm 132 258 534 801

10mm 110 198 402 603

Figure 5.4. 1.56THz Gaussian beam waist size starting at 10mm, propagating 20 meters.


9/13

5.4 Prototype spot scanner for measuring human subjectsThe existing compact radar range uses an outgoing beam to illuminate the target object. Power back-reflecting from thetarget then retraces the illumination beam path and is collected by the receiver. In order to measure human subjects, anew technique was devised and tested, with slight modifications to the existing STL 1.56THz measurement system. Thesimplest approach to measuring a live subject was to focus the 1.56THz terahertz illumination beam onto a spot on the

target subject, and receive any back-reflected power. The process was repeated at fairly high speed across the targetsubject in a raster style scan. The raster style was chosen for the uniform sampling across a rectangular area of interest.

A 10 millimeter diameter spot on a target subject several meters away was the initial goal for the focused beam. UsingGaussian beam calculations, a simple optical configuration was designed. A large scanning mirror was chosen for thedesign. Figure 5.5 shows a schematic of the prototype optical layout.

A prototype system was assembled from available optical and mechanical components at STL. A large mirror, 17 inchesin diameter with a 72 inch focal length was sufficiently large for the experiment to prevent diffractive distortions of the1.56THz beam over the required propagation distance. A two axis computer controlled positioning stage wasassembled, scan control and display software was written in National Instruments LabVIEW. Initially, stepper motorswere used to deflect the two axes on the scanner mechanisms with an angle sufficient to scan a one meter by one meterarea, several meters away. However, the final prototype system enables scans that are much wider than one meter, the

final scan size was approximately 5 meters across by 1-2 meters tall with only a small time penalty for the additionaldata collection.

Major components of the scanner system are shown in Figure 5.5, the beam path is as follows: A beam originates at theterahertz source, expands to fill a focusing mirror, illuminates the planar reflector on the two axis scanner, andpropagates some distance to the target subject (possibly a human subject) located at the beam focus. The terahertz beamreflects off of the subject, retraces backwards through the optical system, and is received by a detector. The down-converted signal is then digitized and displayed on a computer screen in false color, corresponding to intensity of thereturned terahertz energy. For this spot scanning system, the process would repeat for each spot location on the targetsubject, building a complete 2D image.

Figure 5.5. An overview of the optical design, showing relative locations of the monostatic transmitter/receiver, focusing mirror,two-axis scan mirror, and the target subject, initially a mannequin.

In order to be minimally invasive to the already operational and highly complex optical paths of the standard 1.56THzcompact range setup, only simple optical rearrangements were made. Following a few unsuccessful preliminary scans, itwas determined that the system would need to be modified from a slightly bistatic system, where the receiver and


10/13

transmitter have slightly different beam paths, into a strictly monostatic one, where the paths are identical. Theprototype optical design, while simple, was not capable of receiving a beam off the outgoing optical axis.

Following a review by the Universitys Institutional Review Board, the testing of human subjects was considered. Withthe prospect of a scan system illuminating a person with a focused beam from a terahertz laser, acceptable radiationexposure limits for eyes and skin were consulted. The 1.56THz transceivers far-infrared laser output is just over 100

milliwatts. The beam is manipulated through the numerous mirrors, beam splitters, and lenses, and then propagatedapproximately 30 feet, before the focused illumination beam reaches the target subject. The spot size was nominally 10millimeters, and the terahertz beam power measured at the focus location was between 2 and 5 milliwatts, far below therecognized non-ionizing radiation exposure limit of 1000 watts/square meter7. The loss of power down to 2-5 milliwattscan be directly attributed to atmospheric water absorption over the beam pathlength and the need to use several beamsplitters to establish a monostatic path.

6. RESULTS

Because the entire spot scan system was a simple assembly of available hardware integrated with an existing ultrasensitive 1.56THz transceiver, a goal of the project was simply to produce imagery to demonstrate the terahertzphenomenology of a complex target. The time required to collect the imagery was of little concern. Initial scans of a 6

foot tall clothed mannequin were collected from head-to-toe in approximately 7 minutes, with a 2 meter standoffdistance. The unusually short standoff was more of a logistical problem than an optical limitation. In order tounobtrusively retrofit the existing transceiver, a 5 meter standoff was folded, to minimize the modifications to theexisting system. The transmitter beam is sent out in vertical polarization and can be received in either vertical orhorizontal polarization. The preliminary images were novel, but the 10 millimeter spot size, even oversampled in boththe horizontal and vertical directions, was still coarse. Figure 6.1 shows a measurement of a human subject with a 10millimeter spot, 1.56THz beam (right). On the left is a photograph of the subject, taken immediately following theterahertz scan. A concealed handgun is taped to the chest of the subject, underneath a cotton tee shirt. This image wascollected using a vertical polarization transmit beam and a horizontal polarization receiver (VH pol).

Additional modifications were made to the Gaussian optical design, in order to produce a smaller spot size at the target.The initial 10 millimeter spot size was acceptable, but additional resolution was desired. The results of the secondoptical design were quite favorable, with a roughly 6mm spot size and approximately 3 meters of standoff.

The images in figure 6.2 show the actively illuminated, continuous wave (CW), 1.56THz spot scanner image of a livesubject with the reduced terahertz spot size of 6mm. The image on the right is the terahertz image. On the left is thesubject photograph, taken immediately following the scan. A concealed weapon is present on the subject, beneath thecotton shirt. The image also shows a dark L or pistol-like shape, pointing upwards towards the subjects left shoulder.

The terahertz images shown in figures 6.1 through 6.3 are nominally 1000 by 1000 pixels, clearly oversampled for a 5-6millimeter diameter beam. Scan times ranged between 5 and 10 minutes, and were limited solely by the two-axisscanner device. Both images were oversampled by approximately a factor of 10 to 20, mainly due to the scannermechanism in use. Image collection times could easily be reduced by simply reducing the oversampling, and increasingthe speed of the two axis scanner.

Additional terahertz system modifications are under consideration, which could dramatically shorten the time per scan.The tentative goal would be one entire scan, or frame, per second, although multiple frames per second would allow a

more real-time surveillance of human subjects in motion.

The 1.56THz transceiver has the ability to collect co-polarization and cross-polarization imagery. The transmit beamused for all data collections described was vertically polarized. By examining vertical polarization receive (VV), andhorizontal polarization receive (VH), additional image enhancements should be possible. The prototype system was notcapable of collecting both polarizations simultaneously, but measurements were conducted back-to-back switchingbetween V and H receivers.


11/13

Figure 6.1. Imagery collected using the prototype, active illumination, CW, 1.56THz system on the right. Spot size is approximately10 millimeters. On the left is a photograph of the subject taken immediately following the scan. A concealed weapon canbe seen in the terahertz image as an inverted L. Polarization shown is cross-pol (VH).

Figure 6.2. Imagery collected using the prototype, active illumination, CW, 1.56 terahertz system on the right. Spot sizeis approximately 6 millimeters. On the left is a photograph of the subject taken immediately following the scan.A concealed weapon can be seen in the subjects left breast pocket (readers right). Polarization is cross-pol (VH).


12/13

Figure 6.3. 1.56THz co-pol receive(VV) on left, and cross-pol receive (VH) in center. Photograph of the subject is on the right.

Figure 6.3 shows two 1.56THz images of the same human subject and scene, taken several minutes apart. Note thedifferences in the co-polarization terahertz image (left), and the cross-polarization (center). Also notice in figures 6.1,6.2, and 6.3 that the subjects hair appears to rotate the incident polarization more than the surrounding skin. Additionalpolarization effects may prove to be interesting in the exploitation of terahertz imagery of human subjects.

CONCLUSIONS

Two measurement techniques have been demonstrated to produce terahertz imagery of dressed mannequin and humansubjects. A review of two measurement techniques has been described, and imagery was presented at 1.56THz and350GHz. The imagery taken at 1.56THz shows the difficulty in penetrating several layers of cotton material, since thereis little back-reflected power. The 350GHz imagery, on the other hand seems to indicate the possibility of too muchpenetration through the clothing, since there is considerably more back-reflection from the mannequin body.

Effectively, the contrast required to detect the concealed weapon is dramatically reduced in the 350GHz images, asthere is too much backscatter from all objects to be able to discern the weapon from the mannequin body. Reviewingclothing materials transmission spectra, higher transmission occurs at longer wavelengths. Az/El measurementsconducted the 1.56THz and 350GHz showed agreement with the materials spectra. Based on the two measurements at1.56THz and at 350GHz of the clothed mannequin, it would seem that a system designed for the express purpose ofdetecting weapons or contraband under clothing should have a base frequency between 1.5 THz and 350GHz.

A clothed mannequin with a concealed weapon served as an excellent terahertz phenomenology study of a complextarget containing numerous material types. Limitations of the Az/El technique prevent the practical measurements ofhuman subjects, so an alternative scanner prototype was considered.

A spot scanner system was designed and assembled using available optics and mechanical components for a two axisscanner device, and imagery was collected at 1.56THz in a retrofitted compact radar range system. The beam was

directed as a 6mm nominal spot, out to several targets approximately 11 feet away. A two axis scanner directed thefocused spot across the target subject in a raster scan pattern. Reflected terahertz energy was redirected back towards theheterodyne receiver, and image plots were generated using a two dimensional false color display. The scanner consistedof a simple two axis large planar reflector, which was operated a fairly slow speed as a proof-of principle concept forspot scanning a terahertz subject.


13/13

REFERENCES

1. Jason C. Dickinson, Thomas M. Goyette, Jerry Waldman, "High Resolution Imaging using 325GHz and 1.5THzTransceivers", inSession 9: Systems, G. Narayananan, Editor, Proceedings of the Fifteenth International

Symposium on Space Terahertz Technology (STT2004), 373-380, 2004.

2. A.J. Gatesman, A. Danylov, T.M. Goyette, J .C. Dickinson, R.H. Giles, W. Goodhue, J . Waldman, W.E. Nixon, W.Hoen, Terahertz Behavior of Optical Components and Common Materials, inTerahertz for Military and SecurityApplications IV, Proceedings of SPIE Vol. 6212, 2006 (in print)

3. D.L. Mensa, High Resolution Radar Cross-Section Imaging, Artech House, Boston, MA 1991.4. A.J. Gatesman, R.H. Giles, J . Waldman, "High-precision reflectometer for submillimeter wavelengths", JOSA B,

Volume 12, Issue 2, 212-214 February 1995.

5. Ruth M. Woodward, Anthony J. Fitzgerald, Vincent P. Wallace, "Tissue classification using terahertz pulsedimaging",Advanced Characterization, Therapeutics, and Systems XIV, Proceedings of SPIE, Vol. 5318: 23-33

6. Anthony E. Siegman, Lasers, p667, University Science Books, Mill Valley, 19867. Guidelines on Limits of Exposure to Laser Radiation of Wavelengths between 180nm and 1mm, Health Physics

Vol. 71, No. 5, pp 804-819, 1996.

Terahertz imaging of subjects with concealed weapons

Documents